Background: Mammals, globally, are facing population declines. Protecting and breeding threatened populations inside predator-free havens and translocating them back to the wild is commonly viewed as a solution. These approaches can expose predator-naïve animals to predators they have never encountered and as a result, many conservation projects have failed due to the predation of individuals that lacked appropriate anti-predator responses. Hence, robust ways to measure anti-predator responses are urgently needed to help identify naïve populations at risk, to select appropriate animals for translocation, and to monitor managed populations for changes in anti-predator traits. Here, we undertake a systematic review that collates existing behavioural assays of anti-predator responses and identifies assay types and predator cues that provoke the greatest behavioural responses. Methods: We retrieved articles from academic bibliographic databases and grey literature sources (such as government and conservation management reports), using a Boolean search string. Each article was screened against eligibility criteria determined using the PICO (Population–Intervention–Comparator–Outcome) framework. Using data extracted from each article, we mapped all known behavioural assays for quantifying anti-predator responses in mammals and examined the context in which each assay has been implemented (e.g., species tested, predator cue characteristics). Finally, with mixed effects modelling, we determined which of these assays and predator cue types elicit the greatest behavioural responses based on standardised difference in response between treatment and control groups. Review findings: We reviewed 5168 articles, 211 of which were eligible, constituting 1016 studies on 126 mammal species, a quarter of which are threatened by invasive species. We identified six major types of behavioural assays: behavioural focals, capture probability, feeding station, flight initiation distance, giving-up density, and stimulus presentations. Across studies, there were five primary behaviours measured: activity, escape, exploration, foraging, and vigilance. These behaviours yielded similar effect sizes across studies. With regard to study design, however, studies that used natural olfactory cues tended to report larger effect sizes than those that used artificial cues. Effect sizes were larger in studies that analysed sexes individually, rather than combining males and females. Studies that used ‘blank’ control treatments (the absence of a stimulus) rather than a treatment with a control stimulus had higher effect sizes. Although many studies involved repeat measures of known individuals, only 15.4% of these used their data to calculate measures of individual repeatability. Conclusions: Our review highlights important aspects of experimental design and reporting that should be considered. Where possible, studies of anti-predator behaviour should use appropriate control treatments, analyse males and females separately, and choose organic predator cues. Studies should also look to report the individual repeatability of behavioural traits, and to correctly identify measures of uncertainty (error bars). The review highlights robust methodology, reveals promising techniques on which to focus future assay development, and collates relevant information for conservation managers.